Bayesian Learning of Bayesian Networks Bayesian Learning of Bayesian Networks with Informative

This paper presents and evaluates an approach to Bayesian model averaging where the models are Bayesian nets (BNs). Prior distributions are dened using stochastic logic pro-grams and the MCMC Metropolis-Hastings algorithm is used to (approximately) sample from the posterior. Experiments using data generated from known BNs have been con-ducted to evaluate the method. The experiments used 6 dierent BNs and varied: the structural prior, the parameter prior, the Metropolis-Hasting proposal and the data size. Each experiment was repeated three times with dierent random seeds to test the robust-ness of the MCMC-produced results. Our results show that with eective priors (i) robust results are produced and (ii) informative priors improve results signicantly